Why is the SysMemoryUtilization so high on my Amazon Elasticsearch Service cluster?
Last updated: 2020-06-11
I noticed that the SysMemoryUtilization on my Amazon Elasticsearch Service (Amazon ES) cluster is above 90%. Should I be concerned about my system memory utilization usage?
If you see that your system memory utilization is above 90%, it doesn't indicate any heap usage issues or an overloaded Elasticsearch cluster. A system memory utilization above 90% is considered normal, especially for nodes running Amazon ES. Therefore, you do not need to scale up the size of your cluster.
Most of the memory used by Amazon ES is for in-memory data structures. Because Elasticsearch is built on Apache Lucene, Amazon ES uses off-heap buffers for efficient and fast access to files. The Java virtual machine (JVM) also requires some memory.
To determine whether your cluster needs to be scaled up, use an Amazon CloudWatch alarm on these metrics:
- JVM memory pressure
- CPU utilization
- Free storage space
For more information about setting CloudWatch alarms, see Recommended CloudWatch alarms.